The Necessity of Paper Ballots
On November 3, 2020, the citizens of the US will (presumably) hold another national election. COVID-19 notwithstanding, the local, state, and national trajectory has been toward electronic voting. So far, that has largely meant booths at polling stations that electronically tabulate and return results. Longer term, the movement is toward online or mobile voting. Both of these are catastrophic developments for democracy.
I am — like so many people today — heavily reliant on electronics. I use my smart phone for way too much stuff. I use my laptop for even more. Like seriously. I know I should use it less as it opens me up to all kinds of data breaches and hacking and identity theft and so on. Even so, I appreciate and take advantage of the convenience of having technology at my side. I say this only to state from the beginning that I am not some modern-day Luddite, but, rather, someone who really appreciates the convenience of electronic technologies. Furthermore, I am also aware of the digital divide. Even though more and more of our interactions — including the necessities of public life — have moved to a digital platform, far too many citizens do not have access to internet, WiFi, and smart devices. To be clear, I am talking about two connected issues here. One, the shift to electronic booths at in-person polling sites. And two, the shift to online voting. We must resist both of these efforts and insist on paper ballots — either in person or via mail.
Right now, too many citizens cannot (or, can, but only at great cost) get to the polls when they are open. We vote on a work day; we vote on ONE day; we vote in (potentially) inclement weather; and in certain states, politicians have made it harder for working class and working poor voters to get to the polls. They have closed too many polling sites, in targeted neighborhoods populated by people of color. So, allowing citizens to have access to online voting would solve may of these issues — for some. It would not solve any of the problems for people without access.
In the midst of all this, we have seen a push toward more electronic voting machines and electronic vote tabulation. The reasons for this are many. Primarily, the claim is that the results would be faster and more reliable. The evidence is out on these claims. The dangers, as illustrated below, are too great.
So, what are the concerns with electronic voting (in the booth or online)?
Can’t Trust the Coding
I am not a computer programmer. True, I learned some basic coding when I was in high school, but I never did anything else with it. However, based on my very limited knowledge of coding, and based on common sense, we simply cannot trust the coding.
We are told that we can rely on digital voting because (in some localities, anyway), we can walk away with a paper printout of our votes. We can then, the argument goes, rest assured that our votes have been properly cast. Further, in the event of any breakdown, we have a paper trail and the vote can be recounted.
That argument ignores a very simple fact: that printout may not reflect the actual vote. Say we are holding a presidential election that offers Candidate B and Candidate T. Let’s say you walk in to the booth and cast a vote for Candidate B. You take take printout that says you voted for B and you walk away satisfied that you have completed your civic duty. Not so fast.
It would be very easy for a programmer to write a line of code that instructed to machine to print out a receipt that indicated a vote for Candidate B even as it actually cast a vote for Candidate T. Furthermore, it would not have to be every vote, but only an occasional vote. No one would believe that one candidate got 100% of the vote. Instead, the code could instruct the device to effect this move every, say, 200th vote. That small change could alter the outcome of a state’s election.
Let’s look at two examples from the 2016 election. (To be clear, I am NOT saying that this happened in 2016. I’m merely illustrating the potential effects using the last available data.) In that election, 9,091,260 people cast votes in Florida, with Donald Trump receiving 139,770 more than Hillary Clinton (roughly 1.5% of total votes cast). If the voting program altered every 130th vote in favor of Clinton, then Clinton would have won by 95 votes. In Michigan, the margin was even closer, and the shift even more dramatic. There, 4,547,998 votes were cast. Trump won by 11,612 votes (or, .2% of total votes). If the program altered every 200th vote, then it would have switched 22,739 votes, altering the outcome by 45,479 votes. Clinton would have won by 33,867 votes. Or, to imagine the scenario the other way: how few votes would have had to be altered for Trump to win by 11,000 votes? A fraction of one percent.
As you can see, it would not take much to alter the entire outcome of a state, handing all the electoral votes to the new “winner.”
But surely, you say, what company would do such a thing?
Can’t Trust the Creators
According to the US Election Assistance Commission, 19 companies are registered to manufacture and sell voting equipment. Of these 19, Election Systems & Software is the largest, claiming 60% of the market share for all new systems. They have installed state-wide systems in at least 18 US states. As largest, it wields clout. But it is not just ES&S. According to Intelligence Committee member Ron Wyden (D-Oregon), “‘The voting machine lobby, led by the biggest company, ES&S, believes they are above the law. . . . They have not had anybody hold them accountable even on the most basic matters.’”
According to a 2019 article on NBC (authors Popken, McFadden, and Monahan), Election Systems & Software has a few credibility issues. Among the concerns about ES&S is its supply chain through China. As we have seen in recent public battles over Huawei, the concern is that the parts might contain a backdoor that could be exploited by a foreign government such as, in this case, China.
ES&S also has a credibility issue around ownership. As a privately held company, it is not required to disclose owners or revenue. ES&S has, as a result of recent questions from one state Board of Elections, disclosed those who hold more than 5% of the company. But this disclosure was voluntary. We cannot with any certainty know who owns these companies, what their motivations or political allegiances are. Imagine, for a moment, that they primary stakeholder in a polling company is a major donor to either Candidate B or Candidate T. or to the RNC or the DNC. Or is in Vladimir Putin’s inner circle. If they were, could we trust the fairness and integrity of the polling software?
Finally, these companies claim that their code is propriety. They argue that they spent a lot of resources to develop the program, and they own it. To open their code up to inspection or verification, they claim, would jeopardize their intellectual property. They claim that they are protecting trade and company secrets. I understand that in the current climate courts have tended to side with large corporations on privacy issues. Nevertheless, companies like ES&S are paid by the US taxpayers to deliver a voting system for the taxpayers. In that sense, We the People own it. Furthermore, the integrity of the US democratic system is at stake. The code for these systems should be open to inspection — if not by the public, at least by an independent oversight body.
Can’t Trust the Integrity
Beyond the above-mentioned concerns about machine components and ownership, the systems remain vulnerable to hackers. Everything and anything that connects to the web can be hacked. Is being hacked, from people’s pacemakers, to ATMs, to self-driving cars, to CC TV cameras. The Internet of Things is here.
The icing on the cake? At a recent DEFCON, an 11-year old boy hacked into a replica of the Florida elections systems and altered the results in under 10 minutes. Ten minutes. At the same event, an 11-year-old girl hacked the same Florida system in 15 minutes, and tripled the number of votes cast.
I’m sorry, but if TWO 11-year-olds can hack the Florida election system in a matter of minutes, imagine what the professionals in Russia, Syria, China, and North Korea can do.
Digital elections are not secure. They will not be secure any time soon. Hanging chads aside, we need the security of paper ballots.
(I had been thinking about his for a while, and now, of course, it may all be rendered moot by the coronavirus. But we may reach a time when we can go back to polls, and we need to be prepared for that time, too. The push for mail-in ballots is growing, and while it may have some issues surrounding verification, it remains a safer option than going online.)
Ritch Calvin is an Associate Professor of Women’s, Gender, and Sexuality Studies at SUNY Stony Brook. He is the author of Feminist Epistemology and Feminist Science Fiction and editor of a collection of essays on Gilmore Girls.